KR20190004847A - Negative signals for advertisement targeting - Google Patents

Negative signals for advertisement targeting Download PDF

Info

Publication number
KR20190004847A
KR20190004847A KR1020197000504A KR20197000504A KR20190004847A KR 20190004847 A KR20190004847 A KR 20190004847A KR 1020197000504 A KR1020197000504 A KR 1020197000504A KR 20197000504 A KR20197000504 A KR 20197000504A KR 20190004847 A KR20190004847 A KR 20190004847A
Authority
KR
South Korea
Prior art keywords
user
object
topic
negative
associated
Prior art date
Application number
KR1020197000504A
Other languages
Korean (ko)
Inventor
안토니오 필리페 가르시아-마르티네즈
Original Assignee
페이스북, 인크.
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority to US13/566,016 priority Critical patent/US20140040010A1/en
Priority to US13/566,016 priority
Application filed by 페이스북, 인크. filed Critical 페이스북, 인크.
Priority to PCT/US2013/051702 priority patent/WO2014022157A1/en
Publication of KR20190004847A publication Critical patent/KR20190004847A/en

Links

Images

Classifications

    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce, e.g. shopping or e-commerce
    • G06Q30/02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce, e.g. shopping or e-commerce
    • G06Q30/02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
    • G06Q30/0241Advertisement
    • G06Q30/0242Determination of advertisement effectiveness
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce, e.g. shopping or e-commerce
    • G06Q30/02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
    • G06Q30/0241Advertisement
    • G06Q30/0251Targeted advertisement
    • G06Q30/0254Targeted advertisement based on statistics
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce, e.g. shopping or e-commerce
    • G06Q30/02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
    • G06Q30/0241Advertisement
    • G06Q30/0251Targeted advertisement
    • G06Q30/0255Targeted advertisement based on user history
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q30/00Commerce, e.g. shopping or e-commerce
    • G06Q30/02Marketing, e.g. market research and analysis, surveying, promotions, advertising, buyer profiling, customer management or rewards; Price estimation or determination
    • G06Q30/0241Advertisement
    • G06Q30/0251Targeted advertisement
    • G06Q30/0269Targeted advertisement based on user profile or attribute
    • GPHYSICS
    • G06COMPUTING; CALCULATING; COUNTING
    • G06QDATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTING PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q50/00Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
    • G06Q50/01Social networking

Abstract

A user of a social networking system performs actions on various objects managed by a social networking system. Some of these behaviors can indicate that the user has negative feelings about the object. When a social networking system determines that a user performs an action on an object in order to take advantage of such negative emotions when providing content to a user, the social networking system identifies the topic associated with the object and sends a negative emotion to one or more topics Associate. This association between one or more topics and negative emotions can be used to reduce the likelihood that the social networking system will present content on topics related to the user ' s negative emotions.

Description

Negative signals for ad targeting {NEGATIVE SIGNALS FOR ADVERTISEMENT TARGETING}

The present disclosure relates generally to social networking systems, and more particularly to modifying the distribution of content to users of social networking systems based on negative emotions inferred against the user.

A user of a social networking system may establish a connection, relationship or other relationship with other users based on real-life interaction, online interaction, or a combination thereof. Content posted by the user may be made available by the user's connection relationship through one or more of various communication channels within the social networking system, such as a news feed or stream. However, users of social networking systems often receive content that the user is not interested in. In order to improve the content provided to the user including the advertisement, it may be desirable to have a system for inferring topics and other information that the user dislikes in addition to the interests of the user.

In order to enhance the user experience, the social networking system may be able to provide a user's negative < Desc / Clms Page number 2 > with respect to the topic on the content in the social networking system based on negative emotions by other users to the topic in the social networking system, Infer the emotions. Using the inferred emotions, the system selects, filters, predicts, or alters content that is delivered to the user based on the user's inferred negative emotions. For example, a social networking system manages one or more pages containing content for a particular topic, where a particular interaction with a page is known to indicate negative feelings for the topic concerned. In one embodiment, the social networking system associates negative feelings for a topic associated with a page of a particular type of user action performed in connection with the page. When a user of a social networking system interacts with a page, the social networking system deduces that such a user has a negative feel for the topic associated with the page. In addition, a user may interact with other pages within the social networking system (or outside of the social networking system) that are also associated with the same topic, where no (e.g., positive or negative) feelings about the topic associated with another page are known . However, since this user's feelings for the topic have been inferred, the system inferences that interaction with other pages also displays a negative feel for the topic. Thereafter, the system can infer that another user who interacts with the other page also has a negative feel for the same topic. This inferred negative emotion can then be used to generate a negative interest profile that includes negative topics for the user. A user's negative interest profile can be used to perform content filtering, ad targeting, click prediction, or to change the presentation of content to a user.

For example, a social networking system may manage a page entitled " I hate hockey ", and the keyword " hate " in the title refers to a topic (in this example, " hockey " ). ≪ / RTI > Since a set of social networking system users may like the "I hate hockey" page, the social networking system associates the negative feelings of the "hockey" topic with a set of users who like the "I hate hockey" page. If a number of social networking system users who like the "I hate hockey" page also like other pages within the social networking system titled "Hockey?", The social networking system is called "Hockey?" You can infer the negative feelings about the topic of "hockey" for users who like the page. Thus, a user whose social networking system does not identify the feelings for the topic of " hockey " Pages, and social networking systems allow users to "I hate hockey" and "Hockey?" It is inferred that the page has negative feelings for the topic of "hockey" based on the interaction of other users. Thus, the social networking system can add a topic of " hockey " to the user's negative profile, which can then be used to filter out content about the " hockey " to the user.

As discussed above, to improve content distribution, the social networking system may manage a negative profile for a user that includes a topic related to negative emotions. For example, a negative profile may be included or associated with a user's user profile. A negative profile may be used to prevent pages or other content about the topic identified by the blacklist from being presented to the user. This reduces the likelihood of users presenting content to the user with little interest in browsing.

Features and advantages described in the Detailed Description of the Invention are not all inclusive and many additional features and advantages will be apparent to those skilled in the art in view of the drawings, detailed description and claims. In addition, it should be noted that the language used in the detailed description of the present invention has been selected primarily for readability and for purposes of illustration, and is not selected to accurately describe or limit the subject matter of the invention.

Are included in the scope of the present invention.

FIG. 1 illustrates an upper layer block diagram of a system environment for modifying content provided to a user of a social networking system based on a user ' s negative feelings for a content item in accordance with an embodiment.
2 illustrates a flow diagram of a method for providing content to a user based on negative emotions for a content item in a social networking system according to one embodiment.
The Figures illustrate various embodiments of the present invention for purposes of illustration only. Those skilled in the art will readily appreciate that alternative embodiments of the configurations and methods described herein may be utilized without departing from the principles of the invention disclosed herein through the following description.

System structure

1 illustrates a diagram of a system environment for modifying content presented to a social networking system user based on a user ' s negative feelings for a content item in the social networking system 100. As shown in FIG. The user ' s negative feelings for the content item represent a lack of interest of the user to the topic associated with the content item. The user's negative feelings for the topic of the content item with which the user has interacted can be deduced from the interaction of other users of the social networking system 100 with other content items having the same topic as the content item with which the user interacted . The social networking system 100 can deduce negative emotions toward a topic when another social networking system user performs similar interactions with other content items associated with the same topic. Based on the negative feelings for one or more topics, the social networking system 100 may select a content item for the user such that the user is not presented with a content item associated with the topic for which the user has negative feelings. The content items may include any type of media, such as advertisements, coupons, status updates, pages managed by the social networking system 100 or other text messages, location information (e.g., location based on push information) Content. In addition, the social networking system 100 may recommend other users of the social networking system 100 to connect (i.e., become friends) with a particular user based on negative feelings for one or more topics common to the users have.

In general, the social networking system 100 provides the user with the ability to communicate and interact with other social networking system users. As used herein, a " user " may be an individual or an entity (e.g., a business or a third party application). Also, as used herein, a " connection relationship " identifies a user of the social networking system 100 that other users may or may not have established a relationship or other relationship with. The user joins the social networking system 100 and then connects with other users, individuals and entities that the user wants to connect to. The user explicitly adds the connection. For example, the user selects a specific other user who is a friend of the user. Alternatively, the connection between the user and another user may be automatically generated by the social networking system based on the common characteristics of the users (e.g., users who are graduates of the same educational institution). The connections within the social networking system may be bi-directional, or may be only unidirectional. For example, if Bob and Joe are both users and connect together, Bob and Joe are each connection to the other. On the other hand, if Bob tries to connect to Sam to browse the posted content item in Sam, but the Sam does not choose to connect with Bob, a one-way connection can be made where Sam is Bob's connection, . Some embodiments of the social networking system allow a connection to be indirect (e.g., friends of friends) through one or more connection levels.

In addition to interacting with other users, the social networking system 100 provides the user with the ability to perform actions on various types of objects supported by the service. Such objects may include user groups or networks to which a user of a social networking system may belong, events or calendar entries that may be of interest to the user, computer-based applications that the user may use through the service, Or interact with advertisements that a user may perform inside or outside the social networking system. There are several examples of objects that a user may run in the social networking system 100, and many others are possible. While many of the embodiments and examples provided herein relate to specific embodiments of the social networking system 100, other embodiments may also be applicable to other types of social networks, social content, and other types of websites Environment.

User generated content improves the user experience in social networking systems. As described above, the content items may include any type of media content, such as status updates or other text messages, location information, pictures, videos, advertisements and links. A content item is a piece of content represented as an object in the social networking system 100. In this manner, a user of a social networking system is encouraged to communicate with each other by " posting " content items of various types of media over a communication channel to a social networking system. Using the communication channel, the users of the social networking system 100 increase their interaction with each other and participate more frequently in the social networking system. One type of communication channel is a " stream " in which a user is presented with a set of content items that are posted, uploaded, or provided to a social networking system from one or more users of the service. The stream may be updated as the user adds the item of content as a stream. Exemplary communication channels for social networking systems are discussed further in U.S. Patent Application No. 12 / 253,149, filed October 16, 2008, which is incorporated herein by reference in its entirety.

The user interacts with the social networking system 100 using the client device shown in FIG. 1 as user device 105 and connection device 110. User device 105 and / or connection device 110 are for interacting with social networking system 100 and may be any computing device having data processing and data communication capabilities. Examples of client devices include personal computers (PCs), desktop computers, laptop computers, laptops, tablet PCs, personal digital assistants (PDAs), mobile telephones, smart phones or Internet tablets. Such a device may include a camera sensor that allows image and video content to be captured and uploaded to the social networking system 100. Such devices also include a touch screen, gesture recognition system, mouse pad, or other technology that allows a user to interact with the social networking system 100 via a user interface provided by the social networking system 100 .

The interaction between the user device 105, the connection device 110 and the social networking system 100 is typically performed through a network 165, such as the Internet. The network 165 enables communication between the user device 105, the connection device 110, and the social networking system 100. In one embodiment, the network 165 uses standard communication technologies and / or protocols. Thus, the network 165 may be implemented using any of a variety of communication protocols including, but not limited to, Ethernet, 802.11, worldwide interoperability for microwave access (WiMAX), 3G, 4G, LTE, Digital Subscriber Line (DSL), Asynchronous Transfer Mode (ATM), InfiniBand, Advanced Switching), and the like.

In one embodiment, the client device 105 executes a user interface or application that allows a user to interact with the social networking system 100. The user interface allows a user to perform various actions or activities associated with the social networking system 100 and to view information provided by the social networking system 100. Exemplary actions performed using the user interface include adding a connection, posting a message, posting a link, uploading an image or video, updating a user's profile settings, viewing posts, etc. . Examples of information provided by the social networking system 100 that can be viewed using the user interface include: an image or video posted by the user's connection relationship; a comment posted by the user's connection relationship; Messages, wall posts, and the like.

In one embodiment, when user "A" browses data of another user "B", user "A" is referred to as a "viewing user" and user "B" . The user interface allows a viewing user to view general data about news, sports, interests, as well as data of other target users of the social networking system 100. In the user interface, information can be presented to the viewing user in another view. For example, the subject user's social data may be presented to the viewing user as a " profile page " which is an array of the subject's user's social networking data. In addition, the information about the target user may be presented in the form of a news feed including news describing the actions performed by the various target users. In one embodiment, the other views are represented using data and code in a web standard format presented via a browser. For example, the news feed may include any combination of XML, HTML, CSS, JavaScript, plaintext, and Java sent from a server to a web browser running on a client, such as user device 105. In another embodiment, the news feed may include data formatted for presentation via a mobile app or desktop application.

A social network news (or "news") is a collection of data collected by a social networking system 100 configured for display in various social networking system views (user interface views). For example, the news may be presented to a user viewing continuously in a continuously updated real-time news feed in a web browser, in a timeline view, or on a user's profile page. A post set is a collection of one or more posts collected together for display. For example, all news about a particular event, such as a birthday party, can be combined into a news set.

When a user subscribes to the social networking system 100, they may create a user account that allows the user to maintain a permanent and secure identity on the social networking system 100. The user account may include a user profile that stores details or characteristics about the user. Examples of details or characteristics stored in a user profile include name, age, gender, interest, location, education, career, relationship status, and the like. The social networking system 100 may provide a stream of data to the user to notify the user of news and information about the interests of the user as well as to keep the user updated on the activity of the user's connection relationship. Such a stream of data may include news, a collection of related data presented to the user, and a collection of news, a collection of news presented by the user.

The social networking system 100 manages other types of data objects, such as user data objects, behavior objects, and edge objects. The user data store 115 includes a user data object. In one embodiment, the user data object includes user profile information about the user of the social networking system 100. [ For example, the user data object may store characteristics of the user such as, for example, the user's birthday, interest, education, career, photograph of the user, reference to the photograph of the user, or other appropriate user information.

The edge store 120 stores the edge object. In one embodiment, the edge store 120 may store relationships between objects and objects in the social networking system 100 between users, between other users, between users and objects stored in the object store 170 and / And / or edges describing the bonds. Some edges can be defined by the user, which allows the user to specify relationships with other users. For example, a user may create an edge with other users in parallel with the user's actual life relationship, such as a friend, co-worker, partner, and the like. The other edge may allow the user to interact with the social networking system 100, such as expressing interest in pages of the social networking system, sharing a link with other users of the social networking system, and commenting on posts made by other users of the social networking system ) When it interacts with an object. The edge store 120 stores edge objects including information about edges, such as objects, interest and intimacy scores for other users, for example, as described further below.

The action log 125 includes an action performed by a user of the social networking system 100 for a content item or an object for another user stored in the object store 170 or the like. In one embodiment, the action includes information about the interaction that the user performed on the item of content that was logged to enhance the user experience in the social networking system 100. Most of the user ' s most activity towards the content item may be stored as an action in the action log 125. [ For example, the interaction may be a posting of a new comment or a status update, such as removing a content item such as an advertisement or a post, or something simple, such as forming an edge for another user. Additionally, an inaction or lack of activity for a content item may be logged in the activity log 125. For example, if the user does not respond to posts or messages in the social networking system 100, the inactivity may be logged in the activity log 125. [ In one embodiment, each action is assigned to a unique action identifier (ID), and is stored with a user identifier (ID) associated with the user who performed the action on the content item corresponding to the action. The user data included in the user data store 115 and the actions included in the activity log 125 are collectively referred to as narrative data 130. [

The social networking system 100 manages the social graphs that track the relationships between various objects, users, and events captured by the social networking system 100. In the social graph, there are users, user data, and other entities when nodes are connected to each other via an edge. In this embodiment, an edge represents an act of creating a relationship between nodes. For example, a node representing a photo stored in the social networking system 100 may have an edge for the user who uploaded the photo, and such an edge may be an " uploaded by " The same picture may have an edge for some other node representing the user in the picture, and such an edge may be a " tagged in " action. Similarly, a node representing a user in the social networking system 100 may have an edge for each node representing the post the user has made. These edges may all be "posted by" actions. The edges of the social graph may have other types corresponding to other types of actions performed by the user of the social networking system 100.

The social networking system 100 may maintain or calculate a user's " intimacy " measurement for other users (or objects) of the social networking system 100. The intimacy measurement may be expressed as an intimacy score, which may indicate a user's close contact with other users (or objects) of the social networking system 100. The intimacy score of user X for another user Y can be used, for example, to predict whether user X is interested in viewing or viewing the picture of user Y. [ The intimacy score may be calculated by the social networking system 100 through an automated method, including through a predictor function, a machine-learning algorithm, or any other suitable algorithm for determining user friendliness. Since the intimacy scores for various users and objects change over time, the social networking system 100 can store a record of the user's intimacy score history. A system and method for computing user familiarity with other objects in the system as well as with other users of the social networking system 100 is disclosed in U.S. Application No. 12 / 978,265, filed December 23, 2010 , The contents of which are hereby incorporated by reference in their entirety.

In addition, the social networking system 100 includes a user interface manager 135. The user interface manager 135 provides server-side functionality that allows a user of the social networking system 100 to interact with the social networking system 100 using a user interface. When a user requests information from the social networking system 100, the user interface manager 135 may send the requested information in a format that can be displayed via a client device, such as the user device 105 or the connecting device 110, . For example, when a user requests a news feed from the social networking system 100, the user interface manager 135 sends a news and news set to the user device 105 and / or the connected device 110 configured to be displayed on the device . Depending on the type of information requested by the user, the user interface manager 135 may transmit the news, news set, profile page, timeline, or other data to the client device.

The news manager 140 manages the news creation process. The news manager 140 includes a news generator configured to generate news for another purpose (i.e., another view) stored in the news archive 145. The post generator is configured to generate posts for a particular target view, and may limit the selection of narrative data used in post creation based on the target view. For example, the post generator may be configured to generate posts to the photo album view and to limit the narrative data used to generate posts for the narrative data containing or referencing the images. The news generated to be displayed on the user interface may include data that is different from the news generated to be displayed on the desktop computer interface and may be used in other ways to optimize the difference between the desktop computer display and the tactile display, Larger icons for the smartphone screen) can be visually formatted. In addition, the social networking system 100 may include a user who is browsing by a news item including data on a connection relation of a user being browsed, that is, a news containing data on an object user connected with the user who is viewing in the social networking system 100 To limit the news provided to you.

In one embodiment, the news manager 140 generates a news feed that includes a scrollable list of the most relevant news items that the viewing user may be interested in. The interests of the user viewing in the news may be determined by the news manager 140 based on intimacy or other factors. The news manager 140 may generate a timeline, which is a time series news list about a specific target user ordered in time intervals. In some embodiments, the timeline may change the ranking of some news according to other factors such as, for example, social importance or likely engagement value. The news that is configured to display in the timeline is referred to as timeline units. The timeline may also include a special " report " unit that includes a number of timeline units that have been aggregated together. For example, a user may have several wall posts from a friend during November. The user's timeline may include a report unit that includes all posts from a friend for that month. There may be multiple news generators that generate different types of news that are displayed together for news feeds and timelines. A system and method for generating news for a news feed from data captured by a social networking system is disclosed in U.S. Patent Application No. 11 / 503,037 filed on August 11, 2006 and U.S. Patent Application No. 11 / 503,037 filed on August 11, 2006 No. 11 / 502,757, which is incorporated herein by reference in its entirety. The timeline and timeline units are discussed in greater detail in U.S. Patent Application No. 13 / 239,347, filed September 21, 2011, which is also incorporated herein by reference in its entirety.

In one embodiment, the topic extraction engine 150 identifies a topic associated with a content item stored in the object store 170. For example, the topic extraction engine 150 determines one or more topics associated with the content item with which the viewing user interacted. As another example, the topic extraction engine 150 may determine one or more topics associated with the various content items stored by the social networking system 100 in the object store 170. In one embodiment, the topic extraction engine 150 identifies a topic of a content item that is associated with an activity stored in the activity log 125. In order to identify a topic associated with a content item, the topic extraction engine 150 may be configured to perform the following tasks as described further in U.S. Patent Application No. 13 / 167,701, filed June 24, 2011, which is incorporated herein by reference in its entirety: And identify the anchor terms that are described (e.g., in a user's post) with respect to the content items associated with the anchor terms. For example, if an action involves a post or page that includes the text " Go Sharks! &Quot;, the topic extraction engine 150 may: Quot; sharks ", such as " Shark ", " Shark ", " Shark ", and " Loan Shark ". The identified candidate topic represents the potential meaning for the identified anchor term.

In one embodiment, the topic extraction engine 150 removes candidate topics that are determined to be unrelated to anchor terms. For example, the topic extraction engine 150 identifies and analyzes additional terms in an item of content, such as a post, in terms of various identified candidate topics. The topic extraction engine 150 may use a category tree to determine the similarity or degree of relationship between the candidate topics and the terms identified in the content items associated with the behavior. The topic extraction engine 150 may remove one or more candidate topics based on the similarity or relationship diagram received from the category tree.

The topic extraction engine 150 selects the candidate topic from the related candidate topics as the most likely to express the meaning of the anchor term. In one embodiment, the topic extraction engine 150 may be based on the contextual word for the anchor term of the content item associated with the action, based on the user's declared interests associated with the action, and based on the overall context of the action, And generates a score for each candidate topic based on the associated social context. Then, the topic extraction engine 150 selects a candidate topic that expresses the topic for the anchor term based on the generated score. The selected topic is related to the action corresponding to the content item. The topic extraction engine 150 may also infer topics from published videos or pictures represented by behaviors in the activity log 125. [ The topic extraction engine 150 may identify a topic related to the video / photo based on the associated textual metadata describing the content of the video / photo.

In one embodiment, the feedback module 155 identifies the user's negative feelings toward the topic of the content item based on the interaction between the user and the content item. Based on the identified negative feelings, the feedback module 155 generates a negative profile for each user that includes a negative topic for which the user has negative feelings. The feedback module 155 may determine the content to provide to the user using a negative profile associated with the user. In one embodiment, the negative profile for the user serves as a blacklist to identify topics that are not presented to the user. For example, the feedback module 155 may identify content such as an advertisement, post, image, video, news feed or other content item associated with a topic contained in a user's negative profile, Presentation can be prevented. This allows the feedback module 155 to use the user ' s negative feelings for the topic to limit the presentation of content items related to such topics for the user.

The user is able to interact with the content item, but it is unclear whether the user's interaction represents a negative emotion towards the topic associated with the content item. In order to identify a user's negative feelings for the topic associated with the content item with which the user interacts, the feedback module 155 may determine whether the user of the other social networking system, having a negative feel for the topic, Can be inferred. For example, if a user of a social networking system likes or shares a page that expresses negative feelings for a topic, and likes or shares additional pages associated with the same topic, the social networking system may allow the other user, I can deduce that I have negative feelings for. In one embodiment, a user of the other social networking system in which a negative emotion is inferred includes users connected with the user in the social networking system 100. [

For example, if the user interacts with the page " Hockey? &Quot;, it is unclear whether the user has a negative feel for the topic of " hockey. &Quot; However, if a social networking system user interacts with the page "I hate hockey", which also represents a negative emotion for the topic "hockey", and also interacts with "Hockey?", The social networking system is called "Hockey?" Infer that the user interacting with the page has a negative feel for hockey. Thus, the user interaction with " Hockey? &Quot; is used by the social networking system 100 to infer that the user has a negative feel for the topic " hockey ".

In order to determine whether the interaction of the content item and the user is inferred that the user has a negative feel for the topic associated with the content item, the feedback module 115 may determine that the content item < RTI ID = 0.0 > To analyze the interaction with. If another social networking system user having a negative feel for the topic performs an interaction similar to the content item, then the feedback module 115 deduces that the user has a negative feel for the topic based on the user's interaction with the content item . In one embodiment, the feedback module 115 identifies from the action log 125 an interaction by another user with one or more content items that clearly indicate a negative feel for the topic. For example, other users who like pages related to topics that include negative emotional keywords (e.g., dislike, hate, sucks, etc.), and who like the same content item, Is used by the feedback module 115 to infer that the user who likes the item has a negative interest in the topic.

In one embodiment, the feedback module 115 identifies whether a threshold number of other users with negative feelings for the topic has interacted with the content item, and determines whether at least a threshold number of users with negative feelings for the topic Inferring negative emotions from interaction with content items if they worked. Thus, if at least a threshold number of users have a negative feel for the topic of a content item, the interaction of the content item and the user infer that the user also has a negative feel for the topic. The feedback module 115 may add the topic to the user's negative profile if a negative emotion is inferred. In one embodiment, the number of users with negative emotions for a topic can be used to determine a weighting factor for the user ' s negative feelings for the topic. For example, the weighting factor is proportional to the number of users with negative emotions in the topic for a threshold number of users. If the threshold number of users has a negative feel for the topic, a weight of " 1 " can be applied to the user's negative feelings for that topic. However, if half of the users in the critical number have a negative feel for the topic, a weight of " 0.5 " can be applied to the user's negative feelings for that topic. Thus, a sliding scale can be applied to a user's negative feelings for a topic.

In an alternative embodiment, the feedback module 155 identifies an action in the action log 125 that identifies the action the user performed on the content item, in order to identify the user's negative feelings for the topic associated with the content item . The feedback module 155 determines whether the action performed by the user is an act indicative of a negative emotion. For example, the feedback module 155 includes data identifying the type of action associated with the negative emotion and determines whether the action the user performed has the same type identified by the stored data. The specific actions performed by the user on the content items in the social networking system 100 may indicate general negative feelings about the topics or topics corresponding to the content items. For example, closing (i.e., releasing) a content item such as an advertisement, a post, a video, a news feed, a timeline, a news, etc. within a critical time (e.g., one second) And has a negative emotion associated with the topic of the content item. As another example, if a user does not like or hides a content item, it indicates that the user has a negative feel for the topic of the content item. As another example, the textual content posted by the user in the social networking system 100 may be associated with negative connotations indicating negative emotions. The feedback module 155 may identify keywords in textual content, such as " half-off ", " hate ", & For example, a user may create a page in the social networking system 100, such as " I Hate School ", which includes a keyword (e.g., " hate ") associated with a negative emotion that may be associated with a topic relating to the page do.

In addition, a user's interaction with a content item in the social networking system 100 may indicate a negative feel for the topic of the content item. That is, a lack of user behavior on a content item may indicate that the user has a negative feel for the related topic of the content item. For example, a user may receive a content item (e.g., a post, e-mail or message) and a lack of user's response to the content item within the threshold time may indicate a negative feel for the topic associated with the content item. Also, a lack of response from the user may indicate a negative feeling of the user to the user sending the communication to the user, which can be used to change the presentation of the content item to the user from the transmitting user.

Once the feedback module 155 identifies an action performed on the content item indicating a negative emotion, such as an action having a type associated with the negative emotion, from the activity log 125, the feedback module 155 associates the negative emotion with the negative emotion Thereby identifying one or more features of the content item. In one embodiment, the feedback module 155 retrieves one or more topics of the content item determined by the topic extraction engine 150 to associate with negative emotions. Thus, in one embodiment, the feedback module 155 identifies the topic of the content item as causing a negative emotion or related feature. In another embodiment, the feedback module 155 performs linear regression on various features extracted from the content items to identify which features are associated with the negative emotions by the user. The negative feelings of the user for the content item may be stored and related to the action on the content item or content item, or the negative feelings between the user and the topic extracted from the content item are stored.

For example, a user performs an action on an advertisement that indicates negative feelings about the advertisement, such as closing or hiding the advertisement. The feedback module 155 extracts features from the advertisements and performs linear regression on the extracted features. Examples of features extracted from an ad include a landing page for the ad, one or more topics associated with the ad, a page associated with the ad, a sender of the ad, or other feature. The feedback module 155 associates the negative emotions of the user with features of the advertisement (e.g., topics) and stores features associated with the negative emotions of the user. In one embodiment, the feedback module 115 analyzes the actions performed by the user on other objects in the social networking system 100 and the characteristics of the object on which the user performed the action, Analyze the characteristics of the executed object to identify the characteristics of the advertisement. For example, the feedback module 155 performs linear regression based on the characteristics of the object that performed the action to identify the ad feature associated with the negative emotion.

An ad feature associated with a negative emotion may be used to change the advertisement presented to the user later. For example, the advertising feature may be included in a negative interest profile for the user by the feedback module 155 such that a subsequent advertisement with features included in the negative interest profile is not presented to the user. As another example, a negative characteristic of an advertisement may be used in future selection of an advertisement; The expected value of the future advertisement may be attenuated if the advertisement later includes the identified negative characteristic or includes a similar negative characteristic.

In one embodiment, the negative emotions associated with the topic of the advertisement may be provided to an advertiser enabling advertisement insights or adjustment of the indicator to distribute the advertisement to the user. In one embodiment, the feedback module 155 may provide the advertiser with negative emotions (or positive emotions) associated with the topic along with the user's profile information associated with the negative emotions. The advertiser can then determine the characteristics associated with the user with negative emotions associated with the topic, such as age, gender, ethnicity, geographic location, religious beliefs, and the like. Attributes can help advertisers more effectively target ads to users who may be interested in advertising. For example, an advertiser may decide that a male 18-24 year old has negative feelings about the topic of a laundry detergent, and thus may target a user to exclude males 18-24 from receiving a laundry detergent ad .

In one embodiment, the feedback module 155 generates a negative interest profile for another user that includes a topic for which the other user has expressed a negative emotion. A topic associated with a content item for which the user has a negative emotion may be added to the user's negative interest profile when a negative emotion is identified. In one embodiment, the feedback module 155 adds a topic to the user's negative interest profile in response to the user performing at least a threshold number of actions indicating a negative feel for the topic. This prevents the feedback module 155 from incorrectly assigning negative emotions to the topic.

The feedback module 155 may infer the user's negative feelings for additional topics based on the topics included in the negative interest profile. That is, the feedback module 155 may deduce negative feelings for additional topics that are not included in the negative interest profile, based on the relationship or similarity between the additional topic and one or more topics included in the blacklist. For example, if the negative interest profile for the user indicates a negative feel for " cats ", then the feedback module 155 may determine that the cats " cats " &Quot; can be inferred from negative feelings about other topics associated with or related to < RTI ID = 0.0 > A topic whose negative emotions are inferred from a topic in a negative interest profile may then be added to a negative interest profile.

The feedback module 155 enables modification of a content item presented to the user to be changed based on a negative interest profile associated with the user. The candidate content item for the user is identified by the social networking system 100 and the topic or other feature associated with the candidate content item is compared to a negative interest profile associated with the user. The feedback module 155 removes from the candidate content item a subset of the content items having one or more topics that match or are related to the topics included in the negative interest profile. For example, the negative interest profile of the topic may include a topic of " cats ", so the feedback module 155 removes the " cats " or the content item associated with the related topic from the candidate content. In another embodiment, the feedback module 155 may inhibit selection of a content item associated with a topic included in a negative interest profile for the user as a candidate content item. By providing a content item to the user based on the user's negative feelings, the social networking system 100 increases the likelihood that the user will provide the content of interest.

Providing content to users

2 illustrates a method 200 of providing content to a user in a social networking system based on a user's negative feelings. In another embodiment, steps and / or additional steps different from those shown in Fig. 2 may be performed.

The social networking system 100 receives (201) an action performed by a user of the social networking system on one or more objects managed by the social networking system 100. The object may be an advertisement in a social networking system, a post, a news feed, a timeline or any other content item. Examples of actions include closing content items, hiding content items, disliking content, ignoring content, and responding to content. For each object on which the user performed an action, the social networking system 100 identifies (203) a topic associated with the object. In one embodiment, the topic extraction engine 150 in the social networking system identifies (203) one or more topics associated with an object as described in conjunction with FIG. Figure 2 illustrates one embodiment in which one or more topics 203 associated with an object are identified 203, while in another embodiment, the topic extraction engine 150 identifies 203 any suitable feature associated with the object. For example, if the object is an advertisement, the topic extraction engine 150 may identify one or more landing pages associated with the advertisement, a topic associated with the advertisement, an advertiser associated with the advertisement, a keyword associated with the advertisement, 203).

The feedback module 150 determines 205 whether the one or more actions that the user performed on the object are related to the negative feelings for the topic of the object. For example, the feedback module 150 includes a listing of actions associated with negative emotions by the social networking system 100 and determines 205 whether one or more actions performed by the user are included in the list. Examples of behaviors related to negative emotions include: closing (i.e., releasing) an object within a threshold time from being presented, not liking the object, not sending a response to the object within a certain time interval, Providing textual input for an object containing such words or any other suitable action.

If the behavior of one or more of the users performed on the object is related to a negative feel for the topic of the object, then the feedback module 150 deduces that the user has a negative feel for the object. Alternatively, if it is unclear whether the user's behavior on the object is indicative of a negative feel for the topic, the feedback module 150 may be associated with a topic by another social networking system user with a negative emotion known to the topic as described above Negative emotions can be inferred based on interaction with other objects.

In the embodiment illustrated by FIG. 2, the feedback module associates (207) this negative emotion with one or more topics associated with the object. In another embodiment in which the feature extraction module 150 identifies additional features associated with the object, the feedback module 150 associates (207) a negative emotion with one of the identified features. For example, if at least one action performed by a user on an object is associated with a negative emotion, the feedback module 150 associates a negative emotion with the object (207). As another example, the feedback module 150 may determine whether a number of actions performed by the user on the object represent negative emotions, and if the number of actions performed indicating a negative emotion equals or exceeds the threshold, (207).

In one embodiment, the association between the negative emotion and the topic associated with the object is used to generate (209) a negative interest profile. A negative interest profile identifies a topic or other feature associated with a negative emotion by the user. In one embodiment, a negative interest profile is used to select (211) additional content for a user who is later presented 213 to the user. For example, the social networking system 100 compares one or more topics associated with additional content to a negative interest profile, and does not select (211) content associated with at least one topic included in the negative interest profile to present to the user, . As another example, the social networking system 100 may attenuate the expected value of an additional content item associated with a topic included in a negative interest profile, (211).

The foregoing description is directed to inferring a user's negative feelings based on interaction by a user of another social networking system with known negative feelings about the user's interaction with the content item and the topic with the content item associated with the same topic , The foregoing discussion can also be used to infer a user's positive feelings about a topic using an interaction by a user of the social networking system with a positive emotion known to the topic. In addition, other types of emotions can be inferred using the methods described above. Furthermore, while the above description has been described in terms of inferring the user's negative feelings based on the interaction of the content item with the user and by other users in the social networking context, But may be applicable to content items associated with objects managed by the social networking system.

summary

The foregoing description of embodiments of the present invention has been presented for purposes of illustration and is not intended to be exhaustive or to limit the invention to the precise forms disclosed. Those skilled in the art will appreciate that various modifications and changes may be made thereto without departing from the scope of the present invention.

Some portions of the disclosure describe embodiments of the present invention in terms of algorithms or symbolic representations of operations on information. These algorithmic descriptions or representations are commonly used by those of ordinary skill in the data processing arts to convey the substance of their work effectively to those of ordinary skill in the art. These operations, which are functionally, computationally, or logically described, are understood to be implemented by a computer program or equivalent electrical circuit, microcode, or the like. Also, it is often found that the arrangement of such operations can be referred to as modules without loss of generality. The described operations and their associated modules may be implemented in software, firmware, hardware, or any combination thereof.

Any of the steps, operations, or processes described herein may be performed or implemented by one or more hardware or software modules, or by a combination of these and other devices. In one embodiment, a software module is a computer program that includes non-transitory computer readable media including computer program code executable by a computer processor to perform some or all of the described steps, operations, or processes Product.

The embodiments described in the present invention may also be associated with an apparatus for performing the operations herein. The device may include a general purpose computing device that may be specifically configured for the requested purpose and / or selectively activated or reconfigured by a computer program stored on the computer. Such a computer program may be stored in a non-transient computer readable storage medium or any medium suitable for storing electronic instructions that may be coupled to a computer system bus. In addition, any computing system referred to herein may include a single processor, or it may be a structure employing a multiprocessor design for increased computing power.

Finally, the language used herein is theoretically selected for purposes of readability and instruction and is not selected to limit or limit the inventive idea. Accordingly, the scope of the present invention is not to be limited by this detailed description, but is limited only by the scope of the appended claims. Accordingly, the disclosure of embodiments of the present invention is intended to be illustrative, and not to limit the scope of the invention. The scope of the invention is set forth in the following claims.

Claims (22)

  1. Storing a user profile for a first user of the social networking system;
    The method comprising: receiving one or more actions performed by a first user on a first object managed by a social networking system;
    Identifying a topic associated with the first object;
    Retrieving one or more actions previously performed by a second user of the social networking system for one or more second objects also associated with a topic of the first object;
    For a first object that has received at least one action of a first user indicating a positive emotion of the first user for the first object after the first object is displayed to the first user, ≪ / RTI >
    One or more actions previously performed by a second user on a first object that has received one or more actions by a first user, a second user previously performed on one or more second objects that are also associated with a topic of the first object Based on a user profile of a second user indicating one or more actions and a negative emotion for a topic associated with the first object, a first user performs a first action on the first object after the first object is displayed to the first user Inferring that the first user has a negative emotion for a topic associated with the first object, even though the one or more actions indicate that the first user has a positive emotion for the first object;
    Associating the user profile with a topic to store the topic as a negative concern;
    Selecting content to present to a user based at least in part on a negative concern; And
    And presenting the selected content to a user,
    The one or more actions performed by the first user on the first object after the first object is displayed to the first user indicates that the first user has a positive emotion for the first object, One or more actions performed by a user does not indicate whether the first user has a negative emotion or a positive emotion for a topic associated with the first object based on the context of use of the topic in the first object ,
    The second user may be a negative user for a topic associated with the first object that received the one or more actions of the first user that indicated a positive emotion of the first user for the first object after the first object was displayed to the first user Wherein the second user's negative emotions for the topic are based on one or more actions performed by the second user for the second object.
  2. The method according to claim 1,
    The inferring that the first user has a negative emotion for the topic may comprise:
    At least one second user of a social networking system having a user profile displaying at least one action performed by a first user for a topic related to the first object is displayed for at least one second object related to a topic And inferring that the first user has a negative feel for the topic in response to determining to match the previously performed behavior.
  3. The method according to claim 1,
    The inferring that the first user has a negative emotion for the topic may comprise:
    The second users of the social networking system having a user profile in which at least a threshold number of actions performed by the first user exhibited negative emotions for the topic related to the first object were performed previously for one or more second objects related to the topic And inferring that the first user has a negative feel for the topic in response to determining to match the action that was made.
  4. The method according to claim 1,
    The actions previously performed by the second users of the social networking system with the user profile indicating a negative emotion for the topic associated with the first object on the one or more second objects related to the topic may include closing the second object associated with the topic , Disregarding a second object associated with a topic, disregarding a second object associated with a topic, not liking a second object associated with a topic, not sending a response to a second object with respect to a topic within a certain time period, And providing a textual input for a second object related to a topic including one or more words associated with the at least one word.
  5. The method according to claim 1,
    Identifying one or more third objects from which a first user has performed one or more actions; And
    Further comprising associating a negative emotion with one or more third objects based on an action performed on the one or more third objects,
    Wherein the one or more third objects are associated with one or more topics that match one or more topics associated with the first object.
  6. The method according to claim 1,
    Wherein the first object comprises at least one selected from the group consisting of an advertisement, a post, a video, an image, a news, an event and a group.
  7. Receiving a behavior of a user of a social networking system displaying a negative emotion for an object on an object associated with each of the plurality of topics;
    Identifying a topic that is common to all objects from a plurality of topics;
    Determining a total number of behaviors performed by a user who has displayed negative emotions on the object;
    Associating a negative emotion with a topic common to all objects in response to a total number of behaviors associated with negative emotions greater than a threshold; And
    And selecting content to present to a user who performed the action based on the negative feelings about the topic.
  8. 8. The method of claim 7,
    Identifying a negative emotion for an object based on an action on the object includes:
    Determining if one or more actions performed on the object are actions associated with negative emotions by the social networking system; And
    Responsive to a determination that at least one action performed on an object is an affair related to a negative emotional response by the social networking system, identifying a negative emotional response to the object on which at least one action was performed.
  9. 8. The method of claim 7,
    An act associated with a negative emotion by the social networking system comprises an action performed by the social networking system on an additional topic associated with the topic by one or more additional users associated with the negative emotions on the topic.
  10. 9. The method of claim 8,
    Actions related to negative emotions by the social networking system include: closing an object, hiding an object, ignoring an object, not liking an object, not sending a response to an object within a specified time interval, At least one selected from the group consisting of providing text input for an object containing at least one word.
  11. 8. The method of claim 7,
    Behaviors associated with negative emotions by the social networking system include: closing an object, hiding an object, ignoring an object, not liking an object, not sending a response to an object within a specified time interval, At least one selected from the group consisting of providing text input for an object containing at least one word.
  12. 8. The method of claim 7,
    The act involving negative emotions by the social networking system comprises an act performed by the one or more additional users associated with the negative emotions on the topic by the social networking system on an additional topic associated with the topic.
  13. 8. The method of claim 7,
    Selecting content for presentation to a user based on an association between negative emotions for one or more topics comprises:
    Generating a blacklist of users including topics related to negative emotions; And
    Selecting content to provide to a user such that the topic associated with the content is not included in the blacklist based on the blacklist of topics.
  14. 8. The method of claim 7,
    Identifying one or more additional topics relating to a topic associated with a negative emotion; And
    And associating the negative emotions with one or more additional topics.
  15. 8. The method of claim 7,
    Wherein the object includes one or more advertisements, posts, videos, images, news, events, and groups.
  16. 8. The method of claim 7,
    Wherein the content provided to the user is at least one selected from the group consisting of: an advertisement, a video, an image, a post or a link.
  17. Storing a user profile for a user of the social networking system;
    Receiving an action performed by a user on an advertisement and related to a positive emotion for the advertisement by the social networking system after the advertisement is presented to the user;
    Extracting an advertisement feature from the advertisement;
    Identifying, based on the context of use of the advertising features within the advertisement, that the user's action on the advertisement after presentation of the advertisement does not indicate whether the user has a positive emotional or negative emotional response to the ad characteristic;
    Retrieving an action performed by a user for a plurality of additional objects managed by the social networking system and displaying a negative emotion about the additional object after the plurality of additional objects are presented to the user;
    Extracting features from a plurality of additional objects;
    Based on an action performed by a user on a plurality of additional objects indicating a negative emotion for a feature of the identified additional object matching the feature extracted from the advertisement after the plurality of additional objects are presented to the user, Associating advertisement features extracted from the advertisement;
    Storing associations between the ad features and negative emotions in a user profile; And
    Selecting an additional advertisement to present to a user based on an advertising feature associated with a negative emotion.
  18. 18. The method of claim 17,
    Behaviors associated with negative emotions by the social networking system include: closing an object, hiding an object, ignoring an object, not liking an object, not sending a response to an object within a specified time interval, At least one selected from the group consisting of providing text input for an object containing at least one word.
  19. 19. The method of claim 18,
    The step of selecting additional advertisements for presenting to a user based on an advertising feature associated with a negative emotion comprises:
    Selecting an advertisement that does not include an advertisement that does not match an ad feature associated with a negative emotion.
  20. 19. The method of claim 18,
    The step of selecting additional advertisements for presenting to the user based on the selected advertisement features associated with the negative emotions comprises:
    Calculating an expected value associated with each of the plurality of candidate advertisements;
    Reducing an estimate associated with a candidate advertisement that includes an ad feature that matches an ad feature associated with a negative emotion; And
    And selecting an additional advertisement from the plurality of candidate ads based on the calculated expected value.
  21. 19. The method of claim 18,
    The one or more ad features are selected from the group consisting of: a landing page for the ad, one or more topics associated with the ad, a page associated with the ad, and a sender of the ad.
  22. 18. The method of claim 17,
    Wherein the act associated with the negative emotion is based on another action performed on the additional advertisement having one or more ad features matching at least one of the one or more ad features extracted from the ad.
KR1020197000504A 2012-08-03 2013-07-23 Negative signals for advertisement targeting KR20190004847A (en)

Priority Applications (3)

Application Number Priority Date Filing Date Title
US13/566,016 US20140040010A1 (en) 2012-08-03 2012-08-03 Negative signals for advertisement targeting
US13/566,016 2012-08-03
PCT/US2013/051702 WO2014022157A1 (en) 2012-08-03 2013-07-23 Negative signals for advertisement targeting

Publications (1)

Publication Number Publication Date
KR20190004847A true KR20190004847A (en) 2019-01-14

Family

ID=50026393

Family Applications (2)

Application Number Title Priority Date Filing Date
KR20157004801A KR20150038315A (en) 2012-08-03 2013-07-23 Negative signals for advertisement targeting
KR1020197000504A KR20190004847A (en) 2012-08-03 2013-07-23 Negative signals for advertisement targeting

Family Applications Before (1)

Application Number Title Priority Date Filing Date
KR20157004801A KR20150038315A (en) 2012-08-03 2013-07-23 Negative signals for advertisement targeting

Country Status (6)

Country Link
US (1) US20140040010A1 (en)
JP (1) JP6248106B2 (en)
KR (2) KR20150038315A (en)
AU (1) AU2013296857A1 (en)
CA (1) CA2879830A1 (en)
WO (1) WO2014022157A1 (en)

Families Citing this family (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20140172545A1 (en) * 2012-12-17 2014-06-19 Facebook, Inc. Learned negative targeting features for ads based on negative feedback from users
US20140172544A1 (en) * 2012-12-17 2014-06-19 Facebook, Inc. Using negative feedback about advertisements to serve advertisements
JP6122735B2 (en) * 2013-08-28 2017-04-26 ヤフー株式会社 The information processing apparatus, the determination method and determination program
JP5787949B2 (en) * 2013-08-28 2015-09-30 ヤフー株式会社 The information processing apparatus, a specific method and a specific program
US9092742B1 (en) * 2014-05-27 2015-07-28 Insidesales.com Email optimization for predicted recipient behavior: suggesting changes in an email to increase the likelihood of an outcome
US9317816B2 (en) 2014-05-27 2016-04-19 InsideSales.com, Inc. Email optimization for predicted recipient behavior: suggesting changes that are more likely to cause a target behavior to occur
US9088533B1 (en) 2014-05-27 2015-07-21 Insidesales.com Email optimization for predicted recipient behavior: suggesting a time at which a user should send an email
US9563693B2 (en) * 2014-08-25 2017-02-07 Adobe Systems Incorporated Determining sentiments of social posts based on user feedback
US10284537B2 (en) 2015-02-11 2019-05-07 Google Llc Methods, systems, and media for presenting information related to an event based on metadata
US9769564B2 (en) * 2015-02-11 2017-09-19 Google Inc. Methods, systems, and media for ambient background noise modification based on mood and/or behavior information
US10223459B2 (en) 2015-02-11 2019-03-05 Google Llc Methods, systems, and media for personalizing computerized services based on mood and/or behavior information from multiple data sources
US20170177578A1 (en) * 2015-12-18 2017-06-22 Facebook, Inc. Systems and methods for content presentation
JP2018156187A (en) * 2017-03-15 2018-10-04 ヤフー株式会社 Creation device, creation method, and creation program

Family Cites Families (14)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US7346606B2 (en) * 2003-06-30 2008-03-18 Google, Inc. Rendering advertisements with documents having one or more topics using user topic interest
DE102004048552A1 (en) * 2004-10-04 2006-04-13 Nec Europe Ltd. A method for providing information concerning broadcast contents to a user
US20080077494A1 (en) * 2006-09-22 2008-03-27 Cuneyt Ozveren Advertisement Selection For Peer-To-Peer Collaboration
US7730017B2 (en) * 2007-03-30 2010-06-01 Google Inc. Open profile content identification
CA2757668A1 (en) * 2008-04-11 2009-10-15 Desire2Learn Incorporated Systems, methods and apparatus for providing media content
US20100030648A1 (en) * 2008-08-01 2010-02-04 Microsoft Corporation Social media driven advertisement targeting
US9077857B2 (en) * 2008-09-12 2015-07-07 At&T Intellectual Property I, L.P. Graphical electronic programming guide
KR20100079617A (en) * 2008-12-31 2010-07-08 엔에이치엔(주) System of displaying pros and cons on online issue and method thoereof
US20110106630A1 (en) * 2009-11-03 2011-05-05 John Hegeman User feedback-based selection and prioritizing of online advertisements
AU2010333884B2 (en) * 2009-12-23 2014-03-27 Facebook, Inc. Selection and presentation of related social networking system content and advertisements
US20110153412A1 (en) * 2009-12-23 2011-06-23 Victor Novikov Selection and Presentation of Related Social Networking System Content and Advertisements
US20110295612A1 (en) * 2010-05-28 2011-12-01 Thierry Donneau-Golencer Method and apparatus for user modelization
JP5454357B2 (en) * 2010-05-31 2014-03-26 ソニー株式会社 An information processing apparatus and method, and program
US9262517B2 (en) * 2010-08-18 2016-02-16 At&T Intellectual Property I, L.P. Systems and methods for social media data mining

Also Published As

Publication number Publication date
JP6248106B2 (en) 2017-12-13
KR20150038315A (en) 2015-04-08
AU2013296857A1 (en) 2015-02-05
US20140040010A1 (en) 2014-02-06
JP2015531119A (en) 2015-10-29
CA2879830A1 (en) 2014-02-06
WO2014022157A1 (en) 2014-02-06

Similar Documents

Publication Publication Date Title
US9530166B2 (en) Social graph that includes web pages outside of a social networking system
US10116611B2 (en) Prioritizing messages within a message network
US8909569B2 (en) System and method for revealing correlations between data streams
US20120266081A1 (en) Display showing intersection between users of a social networking system
US9377933B2 (en) Displaying social networking system entity information via a timeline interface
US20130013720A1 (en) Scaling Notifications of Events in a Social Networking System
US8521661B2 (en) Suggesting connections to a user based on an expected value of the suggestion to the social networking system
EP2747014A1 (en) Adaptive system architecture for identifying popular topics from messages
US20140136997A1 (en) Targeted advertising based on trending of aggregated personalized information streams
US20140089048A1 (en) Determining Metrics for Groups of Users Defined by Social Signals of a Social Networking System
CA2884201C (en) Customized predictors for user actions in an online system
CA2766132C (en) Gathering information about connections in a social networking service
US9582786B2 (en) News feed ranking model based on social information of viewer
KR101331570B1 (en) Mapping a third-party web page to an object in a social networking system
US8768863B2 (en) Adaptive ranking of news feed in social networking systems
US20120151322A1 (en) Measuring Social Network-Based Interaction with Web Content External to a Social Networking System
US9110953B2 (en) Filtering content in a social networking service
CA2825760A1 (en) Sponsored stories unit creation from organic activity stream
US10264033B2 (en) Selectively providing content on a social networking system
US20110264531A1 (en) Watching a user's online world
US8838581B2 (en) Sending notifications about other users with whom a user is likely to interact
US9082129B2 (en) Providing recommendations on a social networking system page
US8943136B2 (en) Creating and sharing interest lists in a social networking system
US10305847B2 (en) Structuring notification of events to users in a social networking system
US9742822B2 (en) Sponsored stories unit creation from organic activity stream

Legal Events

Date Code Title Description
A107 Divisional application of patent
A201 Request for examination